The goal of this proposal is to develop and evaluate a Variable-Resolution X-ray (VRX) CT scanner capable of varying and increasing its spatial resolution to match the size of any anatomical region under examination. Large anatomical sections (torsos, heads, thighs) will have 2-4 cylmm resolution, medium sections (necks, breasts, extremities, larger research animals) 7-14 cylmm, and small sections (neonatal extremities, fingers, research mice) 24-42 cy/mm. The resolution for pathology specimens and other very small objects will be 50-1 00 cy/mm. In comparison, the maximum resolution of clinical CT is 2 cy/mm. It is anticipated that high- and ultrahigh-resolution clinical CT would assist the detection and assessment of small tumors and tumor morphology in cancer of the head, chest, abdomen, and extremities, as well as in atherosclerosis, arthritis and other joint diseases, bone fractures, abdominal calculi, and many other diseases. Specific applications of high-resolution CT are manifold, including noninvasive evaluation of lung tissue, postoperative evaluation of cochlear implants, analysis of bone architecture and of bone mineral, and endodontic analysis of root canal geometry. Ultrahigh-resolution CT (CT microscopy) has become available and is useful in small-animal cancer research. More recently, CT imaging at the cellular level using synchrotron radiation and large detector assemblies has become possible. The research plan is aimed at developing an advanced version of the VRX CT technology. An optimized (Model 2) VRX CT scanner will be designed, built and evaluated based on information gained from an existing (Model 1) solid-state VRX detector as well as theoretical research and simulations using x-ray Monte Carlo transport programs. In addition, special scanning techniques called "on-the-fly'' imaging and "target reconstruction" will be evaluated. Image quality parameters to be modeled and measured include spatial resolution, noise, contrast resolution and dose, contrast scale, and image artifact production. Evaluations of CT image quality will be made using mechanical and anthropomorphic phantoms, anatomical specimens and animals in vivo. The animal models include C57BLK mice and mongrel dogs. The results of this research may be the development of a new class of clinical CT scanners with greatly enhanced visibility of anatomical detail and hence, significantly improved clinical diagnosis.